A metasomatized 18O-rich veined lithospheric mantle source for ultrapotassic magmas

Wenbo Fan,Neng Jiang,Jun Hu,Danqing Liu,Lang Zhao,Tiejun Li
DOI: https://doi.org/10.1016/j.lithos.2020.105964
IF: 3.5
2021-02-01
Lithos
Abstract:<p>While ultrapotassic rocks are generally considered to be derived from a metasomatized mantle source, whether their characteristically high δ<sup>18</sup>O values represent a primary or a secondary feature and whether the mantle metasomatism is localized or pervasive remain unresolved questions. Here we report that the old, depleted lithospheric mantle in the northern North China craton was locally metasomatized during the late Paleozoic by agents derived from subducted sediments, resulting in incompatible-element-rich phlogopite-bearing pyroxenite veins with anomalously elevated δ<sup>18</sup>O (9.9–12.3‰), high <sup>87</sup>Sr/<sup>86</sup>Sr from 0.7091 to 0.7164, and extremely low ε<sub>Nd</sub> between −25.2 and − 18.2. Subsequent melting of the metasomatic veins plus depleted peridotite host produced the early Mesozoic ultrapotassic rocks with variably enriched isotopic compositions. The proposed scenario can be applied to the Alpine-Himalayan ultrapotassic rocks that are characterized by elevated δ<sup>18</sup>O values and highly evolved SrNd isotopic compositions. It is argued that the metasomatized <sup>18</sup>O-rich source can only be present as veins in the mantle peridotites. The extreme chemical and isotopic variations of the ultrapotassic rocks could reflect those of the metasomatic veins and/or variable proportions of vein material and the peridotite host involved during partial melting. Therefore, the isotopic compositions of mantle-derived rocks may not be directly utilized to infer the nature of the lithospheric mantle. A metasomatized veined mantle source may also be inferred for other types of mantle-derived rocks such as basalts, high-Mg andesites and the Archean sanukitoids.</p>
geochemistry & geophysics,mineralogy
What problem does this paper attempt to address?